Lighting Incompatibility?

[NY]

I'm surprised that it's that way round. I thought the whole advantage of LEDs was that they had higher l/W that fluorescent (and *much* higher than tungsten!). I've learned something!

[Andy Burns]

yep

plenty in the shops over 100l/W including the example I gave from philips, I notice from replacements that the 806 lumen flavour have reduced by a fraction of a watt in the last few years

[The Natural Philosopher]

No, I've always had it in my head that fluoros were the most efficient of all, but were unpleasant. And LEDS were knocking on the door.

And that seems to still be the case. Just

But the difference now is not worth the problems of fluoros. And CFLs were a legislation-driven piece of GreenCrap?.

Next stage is probably OLED panels like those TV backlighting panels.

[The Natural Philosopher]

And plenty of fluoros claim 85l/W

"Lumen is a measure of overall output, and is a scalar quantity, since it is effectively integrated luminosity. Perceived brightness is a physiological function, and is (like all physiological functions) a logarithmic value."

So really it hardly makes a difference

the takeaway condensed version is 'decent fluoro tubes are as bright as LEDs watt for watt'

So that is not a particularly salient issue when deciding which to buy

[The Natural Philosopher]

This is interesting from wiki,

21.5 W LED retrofit for T8 fluorescent tube (230 V) 172 lumens/W 25% eff.

Theoretical limit for a white LED with phosphorescence color mixing

260?300 lumens/W 38.1?43.9% eff.

Now that IS better than fluoros but only a special fluoro replacement LED strip was able to reach the first figure. More conventional LEDS bulbs were < 100.

So it seems things are getting better and there is more efficiency to come, but not *that* much

The low hanging fruit was dumping the 15 lumens per watt incadescents

Its actually a perfect example of the typical Green and wet behind the ears magic thinking

'but LEDS are getting more efficient and soon they wont take any power ay all'

versus -

"even at 100% effiency they will only take ?th the power they do now, and will probably have to be green..."

[Andy Burns]

especially when fluoros emit all the way round, but LEDs tend to shine downwards without needing a reflector.

[Cursitor Doom]

Well then why are you being forced to dump your old fluoros and adopt LEDs then? Is this another EU scam to fleece everyone?

[The Natural Philosopher]

I think he asked if it were worth it, because they were flickering or summat. I'd certainly hang a LED compatible tube in the old fitting, if the fitting was OK.

I like LED light bettter than any other.

[Cursitor Doom]

How does that work, then? I'd have thought the HV gubbins in the tube mounting would zap the LEDs?

I can't see what the issue is. I've no problem at all with CFL and you can get them in a good range of colour temperatures so you don't have to put up with that office-like 'cool white' effect.

[The Natural Philosopher]

Well I am only guessing, but Id say that the same inductor that limits tube current limits the LED current and with enough in series you can make up the voltage, or not, as you prefer. .

CFLs flicker, don't come on for ages, take longer to warm up, then die. How many do you want?

[Cursitor Doom]

On Mon, 21 Dec 2020 20:57:55 +0000, The Natural Philosopher [e-mail] wrote: ?

I'm really struggling to see how the tube supply would be remotely compatible with LEDs. Tubes are voltage operated devices, LEDs current-operated. And as far as I know they still require effective current-limiting to overcome the diode 'knee' effect and thermal runaway.

As many as you've got. I like 'em.

[The Natural Philosopher]

Well no, they are not. Almost exactly the reverse in fact

The thing about a tube is that it is a gas discharge tube, and needs a fair voltage to *strike* , and the assistance of some cathode heaters, but once struck it will draw (almost) unlimited current, at a (constant) voltage IIRC of about 70. The inductor serves to limit the current.

Irrespective of what the voltage across the tube is. You could replace the tube with a dead short and it wouldn't trip anything - the inductor limits the current.

And what that means is that any fairly low impedance load will get the same power delivered. By what amounts to a constant current source comprised of the mains and the inductor. And the fact that the LED and Fluoro efficiencies are similar, means that an LED array fed by such a source will draw virtually the same or similar power and be of equivalent brightness in so doing

I have no idea how that current is apportioned to each LED. Or how many there are. IIRC a whte LED is 4V or so, so to get to no more than 1090V you wouldn't want more than 25 in series, and you would need a bridge rectifier. Or perhaps you could have two chains in opposite polarity, to make use of each half cycle. flickery tho.

I think I'd run a series of rectifiers and caps to drive different chains and put the chains in parallel. Hmm. No that doesn't work ...possibly if enough diodes are in series the average v drop evens out...a you can simply out the chains in parallel

LEDs

That is the inductor doing that to overcome the diode 'knee' effect and thermal

If you think about it the curve of a buncch of LEDS in series and a gas discharge tube are pretty similar.

Next time you are passing then. I do good coffee also

[Cursitor Doom]

These modern lighting LEDs are becoming further and further removed from the small signal devices they evolved from, so I'll bow to your superior knowledge on the subject. I'm very much more electronics-orientated than electricity, anyway. You say these LEDs have a 6V Vf and that is ten times the voltage I would expect to see from the SSD legacy so clearly I'm well out of date with developments in lighting technology!

And I do pretty good advice on tax mitigation. Let's hope we're both still around a year from now, because I can't see myself getting over to England anytime soon, what with the way things are going. :(

[The Natural Philosopher]

I think about 4 for a blue white. There is a relationship between that and the frequency emitted. Quantum crap or summat.

The real point is that the inductor is a lossless current limiter (in theory) and a low loss one in practice. It gets rid of all the volts easily

What's inside the tube electronics wise I am guessing at. Maybe big clive will tear one apart

and that is ten times the voltage I would expect to see

red leds were always a volt and a bit, orange yellow a bit more and green and blue even more

Oh I dunno...

[Cursitor Doom]

Ok, I'll have to read that item later due to shortage of time right now, but I'm harking back over 40 years to the first generation of LEDs which were used for panel illumination on electronic instruments. They'd typically draw about 15mA before going phut and were dimmer than the filament bulbs they were replacing. That's if my memory is not letting me down, that is.

[Cursitor Doom]

Oh, and there were only two colours back then: green and red.

[Paul]

If you plot Vf versus center wavelength, the slope of the line allows derivation of Plancks constant.

(Done for some fool physics lab experiment, with some crusty old spectrometer to check the LEDs. You could still see Plancks fingerprints on the instrument.)

They even have a picture of Mr.Plank here. And you have to plot Va, not Vf. Drat!

*******

The LEDs on my bicycle light (blue with white phosphor) have a Vf of 2.5V at 11mA. The voltage (being a diode curve) rises a bit when you start pumping an amp through the high-power LEDs. I might have a commercial datasheet or two listing 3.2V or so at ampere level. The highest current value I've ever seen anyone put through a single LED was 17 amps. No idea what the voltage is then :-) The LED in that case, was soldered onto a solid copper block, and relied on thermal inertia to protect the LED during a short (blinding) experiment.

LEDs are not very efficient when you do that. You don't get 17x the light when running

17 amps versus 1 amp.

This is one reason the bicycle runs LED arrays and not single LEDs. The LED arrays make better use of the limited power (3W on a good day). Due to the Vf being a bit low, the design only uses 2.5W of the available power. If I could get a 3V LED at 11mA, I'd be using it, as then I could set the operating point at 6V (two LEDs stacked, which is how it always runs). Running

6V @ 500mA would use the available 3W of generator power. Running 5V @ 500mA (with all the LEDs running in parallel in pairs), draws 2.5W.

I was a bit disappointed to see they were running at 2.5V.

Paul

[charles]

I saw my first ones in 1969 - that's over 50 years ago.

[charles]

If both were in the same enclosure, you could have yellow, too.

[Fredxx]

I would suggest a wiring fault. Have you checked if power is getting to the Fluorescent fittings?